U.S. patent application number 13/401780 was filed with the patent office on 2012-08-23 for molecular predictors of therapeutic response to specific anti-cancer agents.
This patent application is currently assigned to The Regents of the University of California. Invention is credited to William J. Gibb, Joe W. Gray, Laura M. Heiser, Wen-Lin Kuo, Anguraj Sadanandam, Paul T. Spellman, Nicholas J. Wang.
Application Number | 20120214829 13/401780 |
Document ID | / |
Family ID | 46653257 |
Filed Date | 2012-08-23 |
United States Patent
Application |
20120214829 |
Kind Code |
A1 |
Spellman; Paul T. ; et
al. |
August 23, 2012 |
Molecular Predictors of Therapeutic Response to Specific
Anti-Cancer Agents
Abstract
Herein is described the use of a collection of 50 breast cancer
cell lines to match responses to 77 conventional and experimental
therapeutic agents with transcriptional, proteomic and genomic
subtypes found in primary tumors. Almost all compounds produced
strong differential responses across the cell lines produced
responses that were associated with transcriptional and proteomic
subtypes and produced responses that were associated with recurrent
genome copy number abnormalities. These associations can now be
incorporated into clinical trials that test subtype markers and
clinical responses simultaneously.
Inventors: |
Spellman; Paul T.;
(Portland, OR) ; Gray; Joe W.; (Lake Oswego,
OR) ; Sadanandam; Anguraj; (Pollachi, IN) ;
Heiser; Laura M.; (Richmond, CA) ; Gibb; William
J.; (San Anselmo, CA) ; Kuo; Wen-Lin; (Lin-Kou
Town, TW) ; Wang; Nicholas J.; (Portland,
OR) |
Assignee: |
The Regents of the University of
California
Oakland
CA
|
Family ID: |
46653257 |
Appl. No.: |
13/401780 |
Filed: |
February 21, 2012 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61444660 |
Feb 18, 2011 |
|
|
|
Current U.S.
Class: |
514/266.4 ;
435/29; 435/6.11; 435/7.24; 506/7; 514/274; 514/300; 514/575 |
Current CPC
Class: |
C12Q 2600/106 20130101;
A61P 35/00 20180101; A61K 31/165 20130101; A61K 31/513 20130101;
C12Q 2600/158 20130101; G01N 33/57484 20130101; A61K 31/167
20130101; A61K 31/517 20130101; C12Q 1/6886 20130101; A61K 31/437
20130101; G01N 2800/52 20130101 |
Class at
Publication: |
514/266.4 ;
514/575; 514/274; 514/300; 506/7; 435/6.11; 435/29; 435/7.24 |
International
Class: |
A61K 31/517 20060101
A61K031/517; A61K 31/165 20060101 A61K031/165; A61K 31/513 20060101
A61K031/513; G01N 33/566 20060101 G01N033/566; A61P 35/00 20060101
A61P035/00; C40B 30/00 20060101 C40B030/00; C12Q 1/68 20060101
C12Q001/68; C12Q 1/02 20060101 C12Q001/02; A61K 31/167 20060101
A61K031/167; A61K 31/437 20060101 A61K031/437 |
Goverment Interests
STATEMENT OF GOVERNMENTAL SUPPORT
[0003] This work was supported in part by Contract No.
DE-AC02-05CH11231 awarded by the Department of Energy, by Grant
Nos. CA058207; U54 CA112970; NHGRI U24, CA126551, and K08CA137153
awarded by the National Cancer Institution of the National
Institutes of Health, and by a Work for Others Agreements
LB06-002417 with GlaxoSmithKline; LB09005492 with Millennium
Pharmaceuticals, Inc.; LB-08004488 with Cytokinetics, Inc.;
LB07003395 with Cellgate, Inc. and LB08005005 with Progen
Pharmaceuticals Ltd. The government has certain rights in the
invention.
Claims
1. A method for identifying a cancer patient suitable for treatment
with an anti-cancer agent selected from the group of vorinostat,
trichostatin A, erlotinib, fluoruracil and GSK1070916 comprising:
(a) measuring the expression level of a target gene in a sample
from the patient; and (b) comparing the expression level of said
gene from the patient with the expression level of the gene in a
normal tissue sample or a reference expression level, wherein an
increase or decrease in the expression level of the target gene
indicates the patient is suitable for treatment with one of the
selected anti-cancer agents.
2. A method for identifying a cancer patient suitable for
treatment, comprising (a) measuring the genomic copy number or
expression level of a gene encoding ER and PR in a sample from the
patient, and (b) comparing the ER and PR genomic copy numbers in
the patient to normal copy number or expression level of the genes
encoding ER and PR, the expression level of the genes encoding ER
and PR in a normal tissue sample or a reference expression level,
or the average expression level of ER and PR in a panel of normal
cell lines or cancer cell lines, wherein a positive level or an
increase in the expression level of ER and PR indicates the patient
is suitable for treatment with vorinostat or trichostatin A.
3. A method for identifying a cancer patient suitable for
treatment, (a) measuring the HER2 protein levels in a sample from
the patient, and (b) comparing the ER and PR protein levels from
the sample to normal ER and PR protein levels in a normal tissue
sample or a reference protein level, or the average protein level
of ER and PR in a panel of normal cell lines or cancer cell lines,
wherein a positive level or an increase in the protein levels of ER
and PR indicates the patient is suitable for treatment with
vorinostat or trichostatin A.
4. A method of treating a cancer patient comprising (a) identifying
a cancer patient who is suitable for treatment with one of five
identified clinical agents, Vorinostat, Trichostatin A, Erlotinib,
Fluoruracil or GSK1070916 and (b) administering a therapeutically
effective amount of the clinical agent.
5. A method of treating a cancer patient comprising (a) obtaining a
biopsy of a cancer patient and identifying the cellular subtype of
the cells in said cancer patient; (b) determining if the subtype is
suitable for treatment with one of five identified clinical agents,
Vorinostat, Trichostatin A, Erlotinib, Fluoruracil or GSK1070916
and (b) administering a therapeutically effective amount of the
clinical agent.
6. The method of claim 5, wherein if the subtype is ER+/PR+, then
the patient is suitable for treatment with Vorinostat and/or
Trichostatin A.
7. The method of claim 5, wherein if the subtype is luminal, then
the patient is suitable for treatment with Vorinostat and/or
Trichostatin A.
8. The method of claim 5, wherein if the subtype is basal, then the
patient is suitable for treatment with Erlotinib and/or
Fluoruracil.
9. The method of claim 5, wherein if the subtype is
ER-/PR-/HER2-Claudin+, then the patient is suitable for treatment
with Erlotinib and/or Fluoruracil.
10. The method of claim 5, wherein if the subtype is Claudin-low,
then the patient is suitable for treatment with GSK1070916 and/or
Fluoruracil.
11. The method of claim 5, wherein if the subtype is KI67+, then
the patient is suitable for treatment with Fluorouracil.
12. The method of claim 5, wherein if the subtype is low or no
20q13 amplification is measured, then the patient is suitable for
treatment with GSK1070916.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Patent
Application No. 61/444,660, filed on Feb. 18, 2011, which is hereby
incorporated by reference in its entirety.
[0002] This application is related to and hereby incorporates by
reference International Patent application no. PCT/US2010/056743,
which is hereby incorporated by reference in its entirety.
BACKGROUND OF THE INVENTION
[0004] 1. Field of the Invention
[0005] The invention relates to the field of biomarkers which are
diagnostic or prognostic for predicting patient response to
specific anti-cancer compounds and therapeutics.
[0006] 2. Background
[0007] The pharmaceutical industry estimates that there are more
than 800 small molecule agents and biologics now under development
for treatment of human malignancies (website for
newmeds.phrma.org). These agents target numerous molecular features
thought to distinguish between tumor and normal cells. These range
from broad specificity conventional therapeutics such as
anti-metabolites and DNA crosslinking agents that currently serve
as mainline breast cancer treatments, to agents that interfere with
aspects to a new generation of agents such as trastuzumab that
selectively target molecular events and pathways that are
deregulated in cancer subsets.
[0008] The general trend in drug development today is toward
development of more targeted agents because these are expected to
show increased efficacy and lower toxicity than conventional
agents. Some drugs such as the ERBB2/EGFR inhibitor, lapatinib show
high target specificity while others such as the SRC inhibitor,
dasatinib, inhibit a broad range of kinases. Given the large number
of agents in clinical development, there is growing recognition
that clinical trials must include predictors of response and must
stratify patients entering the trial.
[0009] Unfortunately, the specificity of most drugs is not
sufficiently high to allow the subtypes in which the drugs will be
effective to be predicted with high confidence. Responsive subsets
can be identified during the course of molecular marker based
clinical trials however this is logistically difficult, expensive
and does not allow experimental compounds to be tested in
subpopulations most likely to respond early in the trials process.
Indeed, the majority of drugs now under development will never be
tested in breast cancer so the probability is high that compounds
that are effective only in subpopulations of breast cancer will be
missed.
SUMMARY OF THE INVENTION
[0010] Personalized medicine promises to deliver specific
treatment(s) to patients likely to benefit from them. Herein it is
shown that testing therapeutic compounds in a panel of breast
cancer cell lines identifies breast cancer subtypes that are likely
to respond to approximately 30% of tested compounds. This supports
the importance of defining response-related molecular subtypes in
breast cancer treatment. It also suggests the utility of
preclinical screening of experimental compounds in
well-characterized cell line panels to identify candidate response
associated molecular signatures that can be used for sensitivity
enrichment in early-phase clinical trials.
[0011] Thus the invention provides for a method for identifying a
cancer patient suitable for treatment with an anti-cancer agent
selected from the group of Vorinostat, Trichostatin A, Erlotinib,
Fluoruracil and GSK1070916 comprising: (a) measuring the expression
level of a target gene in a sample from the patient; and (b)
comparing the expression level of said gene from the patient with
the expression level of the gene in a normal tissue sample or a
reference expression level, wherein an increase or decrease in the
expression level of the target gene indicates the patient is
suitable for treatment with one of the selected anti-cancer
agents.
[0012] A method for identifying a cancer patient suitable for
treatment, comprising (a) measuring the genomic copy number or
expression level of a gene encoding ER and PR in a sample from the
patient, and (b) comparing the ER and PR genomic copy numbers in
the patient to normal copy number or expression level of the genes
encoding ER and PR, the expression level of the genes encoding ER
and PR in a normal tissue sample or a reference expression level,
or the average expression level of ER and PR in a panel of normal
cell lines or cancer cell lines, wherein a positive level or an
increase in the expression level of ER and PR indicates the patient
is suitable for treatment with vorinostat or trichostatin A.
[0013] A method for identifying a cancer patient suitable for
treatment, (a) measuring the HER2 protein levels in a sample from
the patient, and (b) comparing the ER and PR protein levels from
the sample to normal ER and PR protein levels in a normal tissue
sample or a reference protein level, or the average protein level
of ER and PR in a panel of normal cell lines or cancer cell lines,
wherein a positive level or an increase in the protein levels of ER
and PR indicates the patient is suitable for treatment with
Vorinostat or Trichostatin A.
[0014] A method of treating a cancer patient comprising (a)
identifying a cancer patient who is suitable for treatment with one
of five identified clinical agents, Vorinostat, Trichostatin A,
Erlotinib, Fluoruracil or GSK1070916 and (b) administering a
therapeutically effective amount of the clinical agent.
[0015] A method of treating a cancer patient comprising (a)
obtaining a biopsy of a cancer patient and identifying the cellular
subtype of the cells in said cancer patient; (b) determining if the
subtype is suitable for treatment with one of five identified
clinical agents, Vorinostat, Trichostatin A, Erlotinib, Fluoruracil
or GSK1070916 and (c) administering a therapeutically effective
amount of the clinical agent.
[0016] If the subtype is ER+/PR+, then the patient is suitable for
treatment with Vorinostat and/or Trichostatin A. If the subtype is
luminal, then the patient is suitable for treatment with Vorinostat
and/or Trichostatin A. If the subtype is basal, then the patient is
suitable for treatment with Erlotinib and/or Fluoruracil. If the
subtype is ER-/PR-/HER2-Claudin+, then the patient is suitable for
treatment with Erlotinib and/or Fluoruracil. If the subtype is
Claudin-low, then the patient is suitable for treatment with
GSK1070916 and/or Fluoruracil. If the subtype is KI67+, then the
patient is suitable for treatment with Fluorouracil. If the subtype
is low or no 20q13 amplification is measured, then the patient is
suitable for treatment with GSK1070916.
BRIEF DESCRIPTION OF THE FIGURES AND TABLES
[0017] FIG. 1. The cell lines show a broad range of responses to
therapeutic compounds. A. Luminal and ERBB2AMP cell lines
preferentially respond to AKT inhibition. Each bar represents the
response of a single breast cancer cell line to the Sigma AKT1-2
inhibitor, and is colored according to subtype. Cell lines are
ordered by decreasing sensitivity (-log.sub.10(GI.sub.50)). B. Drug
response profiles for compounds with similar mechanisms and targets
are highly correlated. Heatmap shows hierarchical clustering of
correlations between responses of breast cancer cell lines treated
with one of eight compounds. Red indicates positively correlated
sensitivity across the panel of cell lines. Green indicates
anti-correlated drug response profiles. C. Hierarchical analysis of
quantitative responses across cell lines and compounds. Each column
represents one cell line, each row represents median centered
-log10(GI.sub.50) for a particular compound. Both rows and columns
are hierarchically clustered. Only compounds with a significant
subtype effect are included. In the heatmap, red (positive values)
represents sensitivity, green (negative values) represents
resistance, and gray represents missing values. Colored bars below
dendogram indicate sample subtype. Overall, cell lines of similar
subtype tend to cluster together, as do compounds with similar
targets or mechanisms. D. CNAs are associated with compound
response. Boxplots show distribution of response sensitivity for
cell lines with aberrant (A) and normal (N) copy number at the
noted genomic locus. a. 20q13 (STK15/AURKA) amplification is
associated with GSK1070916 (A=7, N=26 samples). b. Amplification at
11q13 (CCND1) is associated with response to carboplatin (A=9, N=28
samples) c. 17q12 (ERBB2) amplification is associated with
sensitivity to BIBW2992 (A=6, N=19 samples), 17-AAG (A=7, N=27
samples), gefitinib (A=7, N=18 samples) and resistance to NU6102
(A=6, N=21 samples).
[0018] FIG. 2. Genomic and transcriptional profiles of the breast
cancer cell lines. A. Hierarchical consensus clustering matrix for
55 breast cancer cell lines showing 3 clusters (claudin-low,
luminal, basal) based on gene expression signatures. For each cell
line combination, color intensity is proportional to consensus. B.
DNA copy number aberrations for 43 breast cancer cell lines are
plotted with log.sub.10(FDR) of GISTIC analysis on the y-axis and
chromosome position on the x-axis. Copy number gains are shown in
red with positive log.sub.10(FDR) and losses are shown in green
with negative log.sub.10(FDR).
[0019] FIG. 3. GI50 calculations are highly reproducible. A. Each
bar represents a count of the frequency of replicated drug/cell
line combinations. Most cell lines were tested only one time
against a particular compound, but some drug/cell line combinations
were tested multiple times. B. Each boxplot represents the
distribution of median absolute deviations for drug/cell line pairs
with 3 or 4 replicates. C. Example drug response curves for HCC1395
treated with cisplatin. Data from three experiments are shown, each
plotted in a unique color. Each dot represents the growth
inhibition following three days of treatment with one of 10
concentrations of cisplatin. For each dose of each experiment,
measurements are performed in triplicate. The x-axis represents
increasing cisplatin concentration; the y-axis indicates growth
inhibition following treatment. A single curve is fit to the set of
30 data points (3 untreated and 27 treated). The vertical line
represents GI50, which is extrapolated from the fitted curve.
Across multiple experimental replicates, the dose-response curve is
highly reproducible. D, E, F. Example drug response curves for
three other cell lines, each treated with a different compound.
Convention as in C.
[0020] FIG. 4. Doubling time varies across cell line subtype. A.
Growth rate, computed as the median doubling time in hours, of the
breast cancer cell lines subtypes are shown as box-plots. The basal
and claudin-low subtypes have shorter median doubling time as
compared to luminal and ERBB2.sup.AMP subtypes, Kruskal-Wallis p
value (p=0.006). B. The ANCOVA model shows strong effects of both
subtype and growth rate on response to 5-FU. Luminal (black) and
basal/claudin-low (red) breast cancer lines each show significant
associations to growth rate but have distinct slopes.
[0021] FIG. 5. The cell line networks are highly significant. The
significance of the subpathways identified by our method was
assessed by comparing the size of our subpathways to the size of
the subpathways generated from a background model in which cells
were randomly partitioned into groups, rather than in the original
subtype definitions. The subpathway sizes were measured in two
ways, the total number of nodes in the subpathway (A,C,E,G) and the
number of nodes in the largest connected component of the
subpathway (B,D,F,H). The luminal (A,B), ERBB2.sup.AMP (C,D),
claudinlow (E,F), and basal (G,H) subpathway sizes are shown as red
dotted lines compared against the distribution of null subpathway
sizes. In all cases the subpathway sizes for the true subtype
partitioning are significantly larger than the subpathway sizes for
the background model.
[0022] FIG. 6A-J provides waterfall plots of breast cancer subtypes
and anti-cancer compounds. Association of clinical subtypes of
breast cancer cell lines with selected anti-cancer compounds are
shown. Each bar represents response sensitivity for one cell line,
cell lines are ordered by sensitivity (-log10(GI50)) and colored to
indicate subtype.
[0023] Table 1. Compounds with significant associations with
specific breast cancer subtypes.
[0024] Table 2. Transcriptional, genomic and phenotypic
characteristics of cell lines in the panel.
[0025] Table 3. Drug response data for each cell line tested
against 77 therapeutic compounds. Data are -log10 transformed.
These data were used to determine subtype specific responses. A tab
delimited .txt file is provided for this table.
[0026] Table 4. Pearson correlations between drug responses for all
compound pairs. A tab delimited .txt file is provided for this
table.
[0027] Table 5. Subtype associations for all therapeutic compounds.
Both raw p-values and FDR-corrected q-values are shown.
[0028] Table 6. Censored drug response data. GI50 values that are
same as maximum experimental concentration used for different drugs
were removed. Data are -log10 transformed. These data were used to
identify responses associated with copy number aberrations. A tab
delimited .txt file is provided for this table.
TABLE-US-LTS-CD-00001 LENGTHY TABLES The patent application
contains a lengthy table section. A copy of the table is available
in electronic form from the USPTO web site
(http://seqdata.uspto.gov/?pageRequest=docDetail&DocID=US20120214829A1).
An electronic copy of the table will also be available from the
USPTO upon request and payment of the fee set forth in 37 CFR
1.19(b)(3).
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
[0029] Preclincial testing in panels of cell lines that mirror
molecular subtypes found in primary tumors promises to allow early
and efficient identification of responsive molecular subtypes as a
guide to early clinical trials. Evidence for the utility of this
approach comes from studies showing that cell line panels predict
responses in (a) lung cancers with EGFR mutations to gefitinib
(Paez J G, et al. (2004) EGFR mutations in lung cancer: correlation
with clinical response to gefitinib therapy. Science
304(5676):1497-1500), (b) breast cancers with ERBB2 amplification
to trastuzumab(Neve R M, et al. (2006) A collection of breast
cancer cell lines for the study of functionally distinct cancer
subtypes. Cancer Cell 10(6):515-527) and/or lapatinib (Konecny G E,
et al. (2006) Activity of the dual kinase inhibitor lapatinib
(GW572016) against HER-2-overexpressing and trastuzumab-treated
breast cancer cells. Cancer Res 66(3):1630-1639), and (c) tumors
with mutated or amplified BCR-ABL to imatinib mesylate (Scappini B,
et al. (2004) Changes associated with the development of resistance
to imatinib (STI571) in two leukemia cell lines expressing p210
Bcr/Abl protein. Cancer 100(7):1459-1471) The NCI's Discovery
Therapeutic Program (DTP) has pursued this approach on large scale
identifying associations between molecular features and responses
to >100,000 compounds in a collection of .about.60 cancer cell
lines (Weinstein J N (2006) Spotlight on molecular profiling:
"Integromic" analysis of the NCI-60 cancer cell lines. Mol Cancer
Ther 5(11):2601-2605; Bussey, K. J. et al. Integrating data on DNA
copy number with gene expression levels and drug sensitivities in
the NCI-60 cell line panel. Mol Cancer Ther 5, 853-867 (2006)).
Although useful for detecting drugs with diverse responses, the
NCI60 panel is arguably of limited power in detecting subtype
specific responses because of the relatively sparse representation
of specific cancer subtypes in the collection. In breast cancer for
example, the collection carries only 6 cell lines. Thus, we have
promoted the use of a collection of .about.50 breast cancer cell
lines for statistically robust identification of associations
between response and molecular subtype in breast cancer. Here we
report the assessment of associations between quantitative growth
inhibition responses and molecular subtypes for 77 compounds
including both FDA approved and investigational agents.
[0030] From a single set of analyses we generated and report here
five of these compounds which are FDA approved agents or compounds
and the molecular subtypes of breast cells which respond to at
least one of the five compounds. Based on the demonstrated
relationship between each clinical agent and a molecularly based
classifier that segregates types of breast cancer cell lines--and
by extension, types of breast tumors--that respond to the agent
from those that do not, herein are described diagnostic or
prognostic methods for determining a patient who would respond
favorably to each of the five compounds and methods and bases for
proposing therapeutic regimens that can be adopted for suitable
patients.
[0031] The five compounds, Vorinostat, Trichostatin A, Erlotinib,
Fluoruracil, GSK1070916, and the molecular subtypes where the five
compounds show preferential activity can be categorized as
follows:
TABLE-US-00001 TABLE 7 Therapeutic Compound Molecular Subtype with
Preferential Activity Luminal (ER+/PR+) Vorinostat, Trichostatin A
Basal (ER-/PR-/HER2-Claudin+) Erlotinib, Fluoruracil Claudin-low
GSK1070916, Fluoruracil KI67+ tumors Fluorouracil Tumors with low
or no 20q13 amplification GSK1070916
[0032] In summary, the following was found that:
[0033] Vorinostat is preferentially active in luminal class cell
lines. This corresponds to ER+/PR+ tumors in clinical studies.
[0034] Trichostatin A is preferentially active in luminal class
cell lines. Again, this corresponds to ER+/PR+ tumors in clinical
studies.
[0035] Erlotinib is preferentially active in basal class cell
lines. This corresponds to the ER-/PR-/HER2-Claudin+ tumor subtype
conventionally described as triple negative.
[0036] The compound identified as GSK.AUR1 (also known as
GSK1070916) by GlaxoSmithKline, is preferentially active against
Claudin-low cell lines, which corresponds to the recently
identified, but rare Claudin-low tumor subtype.
[0037] Fluorouracil (5FU) is active against cell lines with rapid
growth rates. This does not have a molecular correlate at the
present time but might correspond to measure of KI67 staining which
measures growth rate in tumors. Fluoruracil once factoring out
growth rate, is more effective against basal cell lines (both
triple negative and claudin-low) than against luminal cell lines.
This corresponds to the conventional triple negative subset of
breast cancers.
[0038] GSK.AUR1 (GSK1070916) is less effective in tumors that have
genomic DNA copy number amplifications at 20q13, which includes the
AURKA locus which is notably one of the targets of the GSK.AUR1
inhibitor.
[0039] In some embodiments of the invention, a method for
identifying a cancer patient suitable for treatment with an
anti-cancer agent selected from the group of vorinostat,
trichostatin A, Erlotinib, fluoruracil, and GSK1070916, comprising:
(a) measuring the expression level of a target gene in a sample
from the patient; and (b) comparing the expression level of said
gene from the patient with the expression level of the gene in a
normal tissue sample or a reference expression level (such as the
average expression level of the gene in a cell line panel or a
cancer cell or tumor panel, or the like), wherein an increase or
decrease in the expression level of the target gene indicates the
patient is suitable for treatment with one of the selected
anti-cancer agents.
[0040] In one embodiment, the method, further comprising (c)
measuring the genomic copy number or expression level of a gene
encoding ER and PR in a sample from the patient, and (d) comparing
the ER and PR genomic copy numbers in the patient to normal copy
number or expression level of the genes encoding ER and PR, the
expression level of the genes encoding ER and PR in a normal tissue
sample or a reference expression level, or the average expression
level of ER and PR in a panel of normal cell lines or cancer cell
lines, wherein a positive level or an increase in the expression
level of ER and PR indicates the patient is suitable for treatment
with vorinostat or trichostatin A. In another embodiment, the
method, further comprising (c) measuring the HER2 protein levels in
a sample from the patient, and (d) comparing the ER and PR protein
levels from the sample to normal ER and PR protein levels in a
normal tissue sample or a reference protein level, or the average
protein level of ER and PR in a panel of normal cell lines or
cancer cell lines, wherein positive level or an increase in the
protein levels of ER and PR indicates the patient is suitable for
treatment with vorinostat or trichostatin A. Patients identified by
the present invention may also respond to synergistic treatment of
cancer with both vorinostat or trichostatin A.
[0041] In one embodiment, the invention provides for a method of
treating a cancer patient comprising (a) identifying a cancer
patient who is suitable for treatment with one of five identified
clinical agents, Vorinostat, Trichostatin A, Erlotinib,
Fluoruracil, or GSK1070916, and (b) administering a therapeutically
effective amount of the clinical agent. In some embodiments, a
combination of the selected clinical agent and another known
anti-cancer agent, and in other embodiments, the selected clinical
agent and another known anti-cancer agent are administered
concurrently or sequentially.
[0042] In another embodiment, the invention provides for a method
of treating a cancer patient comprising (a) obtaining a biopsy of a
cancer patient and identifying the cellular subtype of the cells in
said cancer patient; (b) determining if the subtype is suitable for
treatment with one of five identified clinical agents, Vorinostat,
Trichostatin A, Erlotinib, Fluoruracil, or GSK1070916, and (b)
administering a therapeutically effective amount of the clinical
agent.
[0043] The present methods describe the measurement and detection
of the expression level of a gene as measured from a sample from a
patient that comprises essentially a cancer cell or cancer tissue
of a cancer tumor. Such methods for obtaining such samples are well
known to those skilled in the art. When the cancer is breast
cancer, the expression level of a gene is measured from a sample
from the patient that comprises essentially a breast cancer cell or
breast cancer tissue of a breast cancer tumor.
[0044] Methods for detection of expression levels of a gene can be
carried out using known methods in the art including but not
limited to, fluorescent in situ hybridization (FISH),
immunohistochemical analysis, comparative genomic hybridization,
PCR methods including real-time and quantitative PCR, and other
sequencing and analysis methods. The expression level of the gene
in question can be measured by measuring the amount or number of
molecules of mRNA or transcript in a cell. The measuring can
comprise directly measuring the mRNA or transcript obtained from a
cell, or measuring the cDNA obtained from an mRNA preparation
thereof. Such methods of extracting the mRNA or transcript from a
cell, or preparing the cDNA thereof are well known to those skilled
in the art. In other embodiments, the expression level of a gene
can be measured by measuring or detecting the amount of protein or
polypeptide expressed, such as measuring the amount of antibody
that specifically binds to the protein in a dot blot or Western
blot. The proteins described in the present invention can be
overexpressed and purified or isolated to homogeneity and
antibodies raised that specifically bind to each protein. Such
methods are well known to those skilled in the art.
[0045] Comparison of the detected expression level of a gene in a
patient sample is often compared to the expression levels detected
in a normal tissue sample or a reference expression level. In some
embodiments, the reference expression level can be the average or
normalized expression level of the gene in a panel of normal cell
lines or cancer cell lines.
[0046] Methods of assaying for ERBB2 or HER2 protein overexpression
include methods that utilize immunohistochemistry (IHC) and methods
that utilize fluorescence in situ hybridization (FISH). A
commercially available IHC test is DAKO HercepTest.RTM. (DAKO
Corp., Carpinteria, Calif.). Patient samples having an IHC staining
score of 0-1,2 is normal, and scores of 2+ may be borerderline,
while results of 2,3+ are scored as positive for multiple copies of
HER2 (HER2 positive).
[0047] A commercially available FISH test is PathVysion.RTM. (Vysis
Inc., Downers Grove, Ill.). The HER2 genomic copy number of a
patient sample is determined using FISH. Generally if a sample is
found to have 3.6 or more copies of HER2 (normal=2 copies), the
patient is determined to be HER2 positive.
[0048] While many HER2-positive patients suffer from metastatic
breast cancer, a patient's HER2 and other tumor cell subtype status
can also be determined in relation to other types of cancers
including but not limited to epithelial cancers such as pancreatic,
lung, cervical, ovarian, prostate, non-small cell lung carcinomas,
melanomas, squamous cell cancers, etc. It is contemplated that the
present methods described herein may find use in prognosis and
predicting patient response to the five compounds that may be used
in various cancer treatments for multiple types of cancers so long
as the patient criteria described herein is present as identifying
a patient suitable for the targeted therapy.
EXAMPLES
Example 1
Identification of Molecular Predictors of Response to 74
Compounds
[0049] The utility of cell lines for identification of clinically
useful molecular predictors of response depends on the extent to
which the diverse molecular mechanisms that determine drug response
are operative in the cell line panel. We have reported previously
on similarities and differences between the cell lines and primary
tumors at the transcription and genome copy number level and we
refine that comparison here using higher resolution platforms.
[0050] The potential clinical utility of these findings is
supported by the fact that in vitro derived molecular predictors of
response to therapeutic compounds are concordant with clinical
results. For example, ERBB2-amplified cell lines are preferentially
sensitive to ERBB2-targeted agents and basal subtype cell lines are
preferentially sensitive to platinum salts, as observed clinically.
That said, additional work remains before the signatures reported
in this study can be used to select patients for clinical trials.
This includes development of robust and reliable molecular assays
that can be applied to clinical samples, establishment of
predictive algorithms with decision making thresholds optimized for
clinical use, and validation of predictive power in multiple
independent studies. To initiate this process, we suggest that the
response associated signatures identified in this study be
developed into standardized assays that can be assessed for
clinical predictive power in early stage clinical trials and used
to design trials that are properly powered to detect the responses
in the clinical subsets predicted by the in vitro studies. Assays
that show positive predictive power in early clinical trials can
then be "locked down" and tested for predictive power in follow-on
clinical trials.
[0051] We anticipate that the power of this in vitro systems
approach will increase as additional molecular features including
mutations, methylation and alternative splicing, are included in
the analysis. In addition, expanding the cell line panel will
increase the power to identify low frequency molecular patterns,
and to develop robust predictive models. Most important, however,
is iterative refinement of the in vitro assay system based on
lessons learned by comparing in vitro predictions with clinical
reality
[0052] Cell line characteristics. Specifically, we used
hierarchical consensus clustering (HCC) of gene expression profiles
to classify 50 breast cancer cell lines and 5 non-malignant breast
cell lines into three transcriptional subtypes: luminal, basal and
the newly described claudin-low (Table 2; PMID 19435916). These
subtypes are related to those described earlier.sup.1 (FIG. 2) but
improved methodology and increased data have refined these classes.
We added a fourth class (ERBB2.sup.AMP) comprised of cell lines
with DNA amplification of ERBB2 to reflect the clinically distinct
treatment category of Her2 positive tumors with the ERBB2-targeted
inhibitors lapatinib and trastuzumab. Finally, a refined high
resolution SNP copy number analysis (FIG. 2B confirms that the cell
line panel models regions of recurrent amplification at 11q13
(CCND1), 17q12 (ERBB2), 20q13 (STK15), or homozygous deletion at
9p21 (CDKN2A)) found in primary tumors. Altogether, this
concordance between cell lines and breast tumors suggests that cell
line subtype mirror the much of the breast tumor diversity found in
patients.
[0053] Drug effects on cell lines. To examine heterogeneity in drug
response across the cell line panel, we assessed quantitative
responses to 77 compounds that are anti-cancer agents across the
cell line panel using a cell growth assay with a quantitative
endpoint measured at three days of continuous exposure to each
agent (Table 3). The anti-cancer agents included clinically
approved agents and compounds still in the product development
cycle with a mix of conventional cytotoxic agents (e.g. taxanes,
platinols, anthracylines) and targeted agents (e.g. SERMs, and
kinase inhibitors). In many cases, several agents targeting the
same protein or molecular mechanism of action were tested. A
variety of response measures were assessed including the
concentration of drug required to inhibit growth by 50%
(GI.sub.50), the concentration necessary to completely inhibit
growth (Total Growth Inhibition, TGI) and the concentration of drug
necessary to reduce the population to 50% of the initial number
(Lethal Concentration 50%, LC.sub.50).
[0054] The design of the assay and the sensitivities of the cell
lines necessitated that even at the highest drug concentrations
tested, for some cell lines one or more of the three responses was
not reached for any given drug. In cases where the underlying
growth data are of high quality, but the end point response (GI50,
TGI, LC50) was not reached the values were set to the highest
concentration tested. GI.sub.50 values represent the lowest
threshold for accurate and diverse data and are the basis for the
remainder of our analysis. GI.sub.50 values were obtained for each
cell line and compound pair that was successfully measured. We
excluded three compounds (PS1145, cetuximab and baicalein) from
further analysis because almost none of the cell lines in the panel
responded strongly.
[0055] A representative waterfall plot showing the variation in
response to the Sigma AKT1-2 inhibitor is shown in FIG. 1A. We
established the reproducibility of the overall data set by making
replicate determinations and median absolute deviation (MAD) of
GI.sub.50 values for 272 drug/cell line combinations with at least
3 replicates. We found that the median MAD was remarkably constant,
just 0.16, regardless of number of replicates (FIG. 2). Response
profiles among the cell lines were generally similar for compounds
targeting similar mechanisms of action (FIG. 1B and Table 4).
[0056] in vitro GI.sub.50 to and clinical relevance. A central goal
of this study was to use the mappings between the breast cancer
cell lines and actual tumors to establish predictors of clinical
response for each. We started our analysis by examining
associations with the four cell line subtypes defined above
(luminal, basal, claudin-low and ERBB2.sup.AMP). The Kruskal-Wallis
test, a non-parametric test, was used to establish associations of
these subtypes with responses to the 74 therapeutic agents.
Overall, 23 of 74 compounds tested and nearly all of the agents
producing strong differential responses across the cell line panel
produced subtype specific responses (p<0.1 after FDR correction
of 222 p-values from all three groups). FIG. 1C shows a
hierarchical clustering of the 26 agents with significant
associations to one or more of the subtypes tested (see also Table
1 and Table 5).
[0057] The top ten most subtype-associated agents were inhibitors
of aspects of receptor tyrosine kinase signaling and histone
deacetylase (Table 1), which had highest efficacy in luminal and
ERBB2.sup.AMP cell lines. Docetaxel, etoposide, and cisplatin
showed preferential activity in basal or claudin-low cell lines,
providing in vitro support for the hypothesis that the standard
chemotherapeutic agents are of benefit to patients with triple
negative or basal-like tumors (PMID 17438091, 20100965). Agents
targeting the mitotic apparatus, including GSK1070916 (AURK B/C
inhibitor), also were more active against basal and/or claudin-low
cell lines.
[0058] Our next effort was focused on using the readily testable
nature of focal high-level copy number aberrations in the clinical
setting, allowing stratification of the patient populations (and
breast cancer cell lines) based on their occurrence. The four
regions of recurrent copy number aberration defined above produced
6 significant associations to single agents (FIG. 1D).
Amplification at 20q13, encoding AURKA, was associated with
resistance to the AURK B/C inhibitor GSK1070916 (Hardwicke M A, et
al. (2009) GSK1070916, a potent Aurora B/C kinase inhibitor with
broad antitumor activity in tissue culture cells and human tumor
xenograft models. Mol Cancer Ther 8(7):1808-1817). This suggests
that amplification of AURKA provides a bypass mechanism for AURK
B/C inhibitors. Amplification at 11q13, encoding CCND1, was
associated with resistance to carboplatin. CCND1 is a G1/S cell
cycle checkpoint gene that monitors for unrepaired DNA damage, and
whose over-expression is known to be associated with cisplatin
resistance in other tumor types (Nakashima T, et al. (2005) The
effect of cyclin D1 overexpression in human head and neck cancer
cells. Eur Arch Otorhinolaryngol 262(5):379-383, Huerta S, et al.
(2003) Gene expression profile of metastatic colon cancer cells
resistant to cisplatin-induced apoptosis. Int J Oncol
22(3):663-670). Amplification at 17q12 (ERBB2) was associated with
sensitivity to BIBW2992 and gefitinib, inhibitors of ERBB2 and/or
EGFR, as well as 17-AAG (HSP90AA1). 17q12 amplification was also
associated with resistance to the CDK1/CCNB1 inhibitor, NU6102,
which may reflect the fact that ERBB2 negatively regulates CDK1(Yu
D, et al. (1998) Overexpression of ErbB2 blocks Taxol-induced
apoptosis by upregulation of p21Cip1, which inhibits p34Cdc2
kinase. Mol Cell 2(5):581-591, Tan M, et al. (2002) Phosphorylation
on tyrosine-15 of p34(Cdc2) by ErbB2 inhibits p34(Cdc2) activation
and is involved in resistance to taxol-induced apoptosis. Mol Cell
9(5):993-1004), thereby diminishing the impact of the CDK1
inhibitor.
[0059] Agent response and other cell line properties. In general,
luminal subtype cell lines grow more slowly than basal or
claudin-low cells (Kruskal-Wallis p=0.006, FIG. 4A) and the range
of doubling times is broad. This raises the possibility that the
cell lines that are most sensitive to the compounds tested are
those that grow most rapidly. If so, then the observed associations
to subtype could represent an association to an obvious covariate.
We tested this hypothesis by assessing the effects of subtype and
doubling time simultaneously using ANCOVA and observed that 20 of
23 compounds had better associations with subtype than with
doubling time (mean log ratio of p-values=0.87, standard deviation
1.09). This identifies subtype membership (or an unmeasured
covariate), rather than doubling time as the major driver of drug
response. Moreover, 11 of 23 subtype specific compounds were more
effective in the more slowly growing luminal cell lines (Table 1),
which would be inconsistent with the hypothesis that only fast
growing cells are easily inhibited. One agent, 5-florouracil , is
not significant in the subtype test alone but shows strong
significance in the ANCOVA model for both class and doubling time.
The response to 5-florouracil decreases as doubling time increases
in both luminal and basal cell lines show this pattern but the mean
sensitivity shifts by approximately one log between the subtypes
(FIG. 4B).
[0060] 1) Vorinostat is preferentially active in luminal class cell
lines. This corresponds to ER+/PR+ tumors in clinical studies.
[0061] 2) Trichostatin A is preferentially active in luminal class
cell lines. Again, this corresponds to ER+/PR+ tumors in clinical
studies.
[0062] 3) Erlotinib is preferentially active in basal class cell
lines. This corresponds to the ER-/PR-/HER2-Claudin+ tumor subtype
conventionally described as triple negative.
[0063] 4) The compound identified as GSK.AUR1 (also known as
GSK1070916) provided to our group by GlaxoSmithKline, is
preferentially active against Claudin-low cell lines, which
corresponds to the recently identified, but rare Claudin-low tumor
subtype.
[0064] 5) Fluorouracil (5FU) is active against cell lines that grow
more quickly than tumors than grow more slowly. This does not have
a molecular correlate at the present time but might correspond to
measure of KI67 staining which measures growth rate in tumors.
[0065] 6) Fluoruracil once factoring out growth rate, is more
effective against basal cell lines (both triple negative and
claudin-low) than against luminal cell lines. This corresponds to
the conventional triple negative subset of breast cancers.
[0066] 7) GSK.AUR1 is less effective in tumors that have genomic
DNA copy number amplifications at 20q13, which includes the AURKA
locus one of the targets of the GSK.AUR1 inhibitor.
[0067] Cell Culture and Nucleotide isolation. Fifty-six breast
cancer cell lines were cultured and nucleotides were isolated as
described previously in Neve R M, et al. (2006) A collection of
breast cancer cell lines for the study of functionally distinct
cancer subtypes. Cancer Cell 10(6):515-527.
[0068] Cell Growth Inhibition Assay. Cells were plated at a density
in 96-well plates such that they would remain in log growth at the
end of assay time. The cells were allowed to attach overnight
before being exposed to drug for 72 h. Compounds were dissolved in
a stock solution of either dimethyl sulfoxide (DMSO) or water, and
a set of 9 doses in 1:5 serial dilution was added in triplicate
wells. The final DMSO concentration in the treated well was 0.3% or
less. The cell growth was determined using Cell Titer Glo assay
(CellTiter-Glo Luminescent Cell Viability Assay, Promega, Madison,
Wis., USA), with slight modification from the manufacturer's
protocol at day 0 (time when drug was added) and day 3 of drug
exposure. Briefly, Cell Titer Glo reagent was diluted with
phosphate-buffered saline (1:1 v:v) and the culture media was
removed from the 96-well plate prior to adding 50 .mu.l per well of
the diluted Cell Titer Glo reagent. Luminescence from the assay was
recorded using BIO-TEK FLx800. From the untreated control wells,
CTG luminescence were measured at day 0 and day 3 (72 hr
later).
[0069] Measurement of Growth rate in cell lines. Doubling time (DT)
was estimated from the ratio of 72 h to 0 h for untreated
wells.
[0070] Analysis of Drug Response Data. Each set of drug response
data consists of measures of the relative amounts of cells still
viable after a sample is subjected to nine 5-fold serial dilutions
of a given drug with 3 replicates each, for a total of 27
observations. A plot of these observations with relative viability
on the y-axis and the log of drug concentrations increasing on the
x-axis suggest a monotonically decreasing curve bounded above and
below on the y-axis. We used a custom-written R package to fit a
curve to the drug response data and calculate a measure of drug
sensitivity.
[0071] Specifically, we used nonlinear least squares to fit these
observations, along with three replicates of the vehicle control
values, with a four-parameter Gompertz curve. Two of the parameters
represent the upper and lower asymptotes of the curve, and the
other two adjust the slope and point of inflection. We used a
Gompertz model because it allows for flexibility and asymmetry
about the inflection point. The fitted curve for each set is then
transformed into a GI curve, using the method described by the
NCI/NIH DTP Human Tumor Cell Line Screen Process (Russ, A. P. &
Lampel, S., The druggable genome: an update. Drug Discov Today 10
(23-24), 1607-1610 (2005)) and as previously described in Monks, A.
et al., Feasibility of a high-flux anticancer drug screen using a
diverse panel of cultured human tumor cell lines. J Natl Cancer
Inst 83 (11), 757-766 (1991). The percent growth curve is
calculated as [(T-T0)/(C-T0)].times.100, where T0 is the cell count
at day 0, C is the vehicle control (for example 0.3% DMSO without
drug) cell count at day 3, and T is the cell count at the test
concentration. The GI50 value is determined as the drug
concentration that results in 50% growth at 72 h drug exposure.
[0072] We filtered the drug response data on four quality control
metrics: 1) median standard deviation across the 9 concentrations
less than 0.20; 2) doubling time within 2 standard deviations of
the median doubling time for a particular cell line; 3) slope of
the fitted Gompertz curve to be greater than 0.25; 4) growth
inhibition at the maximum concentration less than 50% for cell
line/drug combinations with no clear response. Approximately 80% of
the drug plates pass all filtering requirements.
[0073] SNP Array Processing and DNA Copy Number Analysis.
Affymetrix Genome-Wide Human SNP Array 6.0 quality and data
processing was performed using the R statistical framework
(R-project website) based aroma.affymetrix6. The breast cancer cell
line SNP arrays were normalized using 20 normal sample arrays as
described in Bengtsson, H., Irizarry, R., Carvalho, B., &
Speed, T. P., Estimation and assessment of raw copy numbers at the
single locus level. Bioinformatics (Oxford, England) 24 (6),
759-767 (2008). The raw copy number for each sample obtained from
aroma.affymetrix were segmented using circular binary segmentation
(CBS) algorithm using R and Bioconductor (Gentleman, R. C. et al.,
Bioconductor: open software development for computational biology
and bioinformatics. Genome biology 5 (10), R80 (2004)) based
DNAcopy (Olshen, A. B., Venkatraman, E. S., Lucito, R., &
Wigler, M., Circular binary segmentation for the analysis of
array-based DNA copy number data. Biostatistics (Oxford, England) 5
(4), 557-572 (2004)). The significant DNA copy number changes were
analyzed using MATLAB based Genomic Identification of Significant
Targets in Cancer (GISTIC) as described in Beroukhim, R. et al.,
Assessing the significance of chromosomal aberrations in cancer:
methodology and application to glioma. Proceedings of the National
Academy of Sciences of the United States of America 104 (50),
20007-20012 (2007).
[0074] Drug screening. Each drug included in the statistical
analysis satisfied the following screening criteria for data
quality: [0075] Missing values--No more than 40% of GI50 values can
be missing across the entire set of cell lines. [0076]
Variability--For at least 3 cell lines, either
[0076] GI50>1.5GI50.sub.median, or
GI50<0.5GI50.sub.median
[0077] where GI50.sub.median is the median GI50 for a given drug.
Any compounds failing these criteria were excluded from the
statistical analysis. Source code for the screening algorithm is
included with Supplementary Information.
[0078] Exon array processing. Gene expression data for the cell
lines were derived from Affymetrix GeneChip Human Gene 1.0 ST exon
arrays. Gene-level summaries of expression were computed using the
aroma.affymetrix R package (Bengtsson et al, 2008), with quantile
normalization and a log-additive probe-level model (PLM) based on
the HuEx-1.sub.--0-st-v2,DCCg, Spring 2008 CDF. Transcriptional
profiles derived from the Affymetrix exon arrays have been shown to
accord well with those derived from Affymetrix HG-U133 Plus 2.0
arrays (Pradervand et al, 2008). Transcript identifiers were
converted to HGNC gene symbols by querying the Ensembl database
using the BioMart R package. The resulting expression profiles were
subsequently filtered to capture only those genes expressing a
standard deviation greater than 1.0 on the log.sub.2-scale across
all cell lines.
[0079] Consensus clustering. Cell line subtypes were identified
using hierarchical consensus clustering (Monti et al, 2003).
Consensus was computed using 500 samplings of the cell lines, 80%
of the cell lines per sample, agglomerative hierarchical clustering
and average linkage. R source code is included with Supplementary
Information.
[0080] Merging of Microarray Datasets. A gene expression microarray
dataset (GSE10885) containing breast tumors with all the five
breast cancer subtypes and metaplastic breast tumors2 were obtained
from Gene Expression Omnibus (GEO) 10. Breast cancer cell line and
breast tumor gene expression profiles were screened by selecting
gene symbols with standard deviation (SD)>0.8. The merging of SD
selected datasets was performed using DWD as described 11,12. Each
dataset was column (samples) normalized to N(0,1) and row (genes)
normalized by median centering. The processed datasets were merged
using Java base DWD (Benito, M. et al., Adjustment of systematic
microarray data biases. Bioinformatics (Oxford, England) 20 (1),
105-114 (2004)) and finally, median centered across row (genes). HC
of the merged dataset was performed using Cluster.
[0081] Associations of Subtype and Response to Therapeutic Agents.
Associations between drug response and subtype were assessed for:
(a) luminal vs. basal vs. claudin-low; (b) luminal vs.
basal+claudin-low; and (c) ERBB2-AMP vs. non-ERBB2-AMP. Differences
between -log10(GI.sub.50) of the groups were compared with a
non-parametric Kruskall-Wallis ANOVA. The p-values for the three
sets of tests were combined and the Benjamini-Hochberg False
Discovery Rate (FDR q-value) was used to correct for multiple
testing. For the three-sample test, the most sensitive group was
identified by performing a post-hoc analysis on the significant
compounds in which we compared each group to all others. The
p-values for the post-hoc test were adjusted together. In all
cases, q<0.10 was deemed significant. If the basal +claudin-low
group was significant in scheme 2, but only one of these groups was
significant in scheme 1, precedence was given to the 3 sample case
when assigning class specificity. There was no minimum difference
in medians required.
[0082] Association of Growth Rate and Response to Therapeutic
agents. We performed a 2-way ANCOVA to assess the effects of cell
line class and growth rate on drug sensitivity. Specifically, we
fit a linear model that looks for a separate regression line for
each class of cell lines:
GI50=class+growth rate+error
[0083] We performed a separate ANCOVA for each of the three cell
line classification schemes, which yielded 6 sets of p-values (2
main effects.times.3 classification schemes). We used a single FDR
correction to assess significance, and declared FDR
p-values<0.20 to be of interest. We performed these analyses in
R with the functions lm and Anova, which is available as part of
the car package.
[0084] Assessment of GI50 replicates. We used the median absolute
deviation (MAD) to assess the reliability of our replicate measures
of GI50. The MAD is a measure of deviation, similar to, but more
robust than the standard deviation. We computed the MAD as a
function of number of replicates for each drug/cell line
combination with more than 3 replicates.
[0085] Association of Genomic Changes and Response to Therapeutic
Agents. A t-test was used to assess the association between
recurrent copy number changes at 9p21, 11q13, 17q12 and 20q13, as
identified in the GISTIC analysis, and drug response. Cell lines
with low or no amplification were combined into a single group and
compared to cell lines with high amplification. A similar analysis
was performed for regions of deletion. Cell lines for which the
GI.sub.50 was equal to the maximum concentration tested were
omitted from analysis (e.g., after censoring lapatinib, there were
only 2 samples in the amplified copy number group for 17q12; Table
S6). Compounds were omitted if the distribution deviated greatly
from normality, as assessed by QQ plot. The complete set of
p-values was adjusted for multiple comparisons, and q .ltoreq.0.10
was deemed significant.
[0086] Identification of subtype pathway markers Interconnected
genes that collectively showed differential IPLs with respect to
subtype were identified by treating each subtype as a
dichotomization of the cell lines into a group containing the
subtype of interest and a group containing the remaining cell
lines. The R implementation of the two-class Significance Analysis
of Microarrays (SAM) algorithm (Tusher V G, Tibshirani R, & Chu
G (2001) Significance analysis of microarrays applied to the
ionizing radiation response. Proc Natl Acad Sci USA
98(9):5116-5121) was used to compute a differential activity (DA)
score for each concept in the SuperPathway. For subtypes, positive
DA corresponds to higher activity in the subtype compared to the
other cell lines.
[0087] Integration of copy number and transcription measurements
identifies biologically relevant SuperPathways. We used the network
analysis tool PARADIGM (Vaske C J, et al. (Inference of
patient-specific pathway activities from multi-dimensional cancer
genomics data using PARADIGM. Bioinformatics 26(12):i237-245) to
identify pathway based mechanisms that underlie subtype specific
responses. PARADIGM uses copy number and transcription data to
calculate integrated pathway levels (IPLs) for 1441 curated signal
transduction, transcriptional and metabolic pathways (see
Kristensen, et al, this issue). We compared IPLs for cell lines and
primary breast tumors using data from The Cancer Genome Atlas
(TCGA) project (Website for cancergenome.nih.gov), and found a
general concordance between transcriptional subtype and pathway
activity across the two cohorts (data not shown). This subtype
specific pathway activity likely explains much of the observed
subtype specific responses.
[0088] SuperPathway analysis of differential drug response among
the cell lines also revealed subnet activities that provide
information about mechanisms of response. For example, basal cell
line sensitivity to the DNA damaging agent, cisplatin, was
associated with up-regulation of a DNA-damage response subnetwork
that includes ATM and CHEK1, key genes associated with response to
cisplatin (Siddik Z H (2003) Cisplatin: mode of cytotoxic action
and molecular basis of resistance. Oncogene 22(47):7265-7279) (data
not shown). Likewise, ERBB2.sup.AMP cell line sensitivity to
geldanamycin (HSP90 inhibitor) was associated with up-regulation of
an ERBB2-HSP90 subnetwork (data not shown). This is consistent with
the known ERBB2 degradation induced by geldanamycin binding
(Blagosklonny M V (2002) Hsp-90-associated oncoproteins: multiple
targets of geldanamycin and its analogs. Leukemia 16(4):455-462;
Baselga J & Swain S M (2009) Novel anticancer targets:
revisiting ERBB2 and discovering ERBB3. Nat Rev Cancer
9(7):463-475).
Example 2
Identifying Patient Response to One of Five Compounds
Identified
[0089] Vorinostat, trichostatin A, Erlotinib, and fluoruracil are
currently approved for use in patients with various cancers. For
example, patients eligible for erlotinib or fluorouracil therapy
would be triple negative (ER-/PR-/HER2-Claudin+) patients. Paraffin
embedded tumor blocks from patient biopsy could be assessed for
KI67+ staining using standard molecular approaches. If positive for
KI67 staining, then the patient should be prescribed
fluoruracil.
[0090] On the other hand, patients with cancers found to be ER+/PR+
would be instead prescribed vorinostat or trichostatin A. Thus,
determining the patient response profile will eliminate therapies
to patients where response is predicted to be resistant.
REFERENCES
[0091] 1 Neve, R. M. et al., A collection of breast cancer cell
lines for the study of functionally distinct cancer subtypes.
Cancer cell 10 (6), 515-527 (2006).
[0092] 2 Hennessy, B. T. et al., Characterization of a naturally
occurring breast cancer subset enriched in
epithelial-to-mesenchymal transition and stem cell characteristics.
Cancer research 69 (10), 4116-4124 (2009).
[0093] 3 Serrano, M., Hannon, G. J., & Beach, D., A new
regulatory motif in cell-cycle control causing specific inhibition
of cyclin D/CDK4. Nature 366 (6456), 704-707 (1993).
[0094] 4 Russ, A. P. & Lampel, S., The druggable genome: an
update. Drug Discov Today 10 (23-24), 1607-1610 (2005).
[0095] 5 Monks, A. et al., Feasibility of a high-flux anticancer
drug screen using a diverse panel of cultured human tumor cell
lines. J Natl Cancer Inst 83 (11), 757-766 (1991).
[0096] 6 Bengtsson, H., Irizarry, R., Carvalho, B., & Speed, T.
P., Estimation and assessment of raw copy numbers at the single
locus level. Bioinformatics (Oxford, England) 24 (6), 759-767
(2008).
[0097] 7 Gentleman, R. C. et al., Bioconductor: open software
development for computational biology and bioinformatics. Genome
biology 5 (10), R80 (2004).
[0098] 8 Olshen, A. B., Venkatraman, E. S., Lucito, R., &
Wigler, M., Circular binary segmentation for the analysis of
array-based DNA copy number data. Biostatistics (Oxford, England) 5
(4), 557-572 (2004).
[0099] 9 Beroukhim, R. et al., Assessing the significance of
chromosomal aberrations in cancer: methodology and application to
glioma. Proceedings of the National Academy of Sciences of the
United States of America 104 (50), 20007-20012 (2007).
[0100] 10 Edgar, R., Domrachev, M., & Lash, A. E., Gene
Expression Omnibus: NCBI gene expression and hybridization array
data repository. Nucleic acids research 30 (1), 207-210 (2002).
[0101] 11 Benito, M. et al., Adjustment of systematic microarray
data biases. Bioinformatics (Oxford, England) 20 (1), 105-114
(2004).
[0102] 12 Herschkowitz, J. I. et al., Identification of conserved
gene expression features between murine mammary carcinoma models
and human breast tumors. Genome biology 8 (5), R76 (2007).
[0103] All patents, patent applications and references made herein
are hereby incorporated by reference in their entirety for all
purposes.
TABLE-US-00002 TABLE 1 Therapeutic compounds that show significant
subtype-specificity. Each column represents q-values for one ANOVA.
Compounds are ranked by the minimum q-value achieved across the
three tests. Basal/Claudin- Basal + Claudin- ERBB2AMP/not Compound
Target low/Luminal low/Luminal ERBB2AMP Subtype specificity
Lapatinib EGFR, ERBB2 7.23E-02 3.34E-02 2.26E-06 Luminal/ERBB2AMP
Sigma AKT1-2 inh. AKT1, AKT2 1.17E-03 2.63E-04 1.29E-01 Luminal
GSK2126458 PIK3C A/B/D/G 1.27E-03 1.27E-03 8.67E-02
Luminal/ERBB2AMP Gefitinib EGFR 4.89E-01 3.35E-01 4.14E-03 ERBB2AMP
BIBW 2992 EGFR, ERBB2 6.93E-01 8.08E-01 6.39E-03 ERBB2AMP
GSK2119563 PIK3CA 2.85E-02 8.11E-03 8.67E-02 Luminal/ERBB2AMP
Rapamycin MTOR 1.45E-02 8.11E-03 3.84E-01 Luminal AG1478 EGFR
9.34E-01 9.34E-01 2.60E-02 ERBB2AMP Etoposide TOP2A 3.34E-02
5.13E-02 8.89E-01 Claudin-low LBH589 HDAC 5.14E-02 3.34E-02
3.22E-01 Luminal Vorinostat HDAC 7.23E-02 3.34E-02 6.89E-01 Luminal
Cisplatin DNA cross- 8.45E-02 4.31E-02 8.52E-01 Basal/Claudin-low
linker Fascaplysin CDK4 4.83E-02 4.31E-02 3.70E-01 Luminal
Docetaxel TUBB1, BCL2 8.67E-02 4.83E-02 8.44E-01 Basal/Claudin-low
GSK1070916 AURK B/C 5.13E-02 4.83E-02 4.82E-01 Claudin-low PD173074
FGFR3 5.13E-02 3.68E-01 5.06E-01 Claudin-low Trichostatin A HDAC
1.22E-01 5.13E-02 7.10E-01 Luminal Triciribine AKT, ZNF217 8.67E-02
5.91E-02 3.56E-01 Luminal CGC-11047 Polyamine 6.51E-02 1.25E-01
8.08E-01 Basal analogue Temsirolimus MTOR 1.64E-01 7.25E-02
1.29E-01 Luminal VX-680 AURK A/B/C 2.95E-01 4.02E-01 7.77E-02 not
ERBB2AMP 17-AAG HSP90AA1 1.83E-01 1.10E-01 8.67E-02 ERBB2AMP
Erlotinib EGFR 9.48E-02 2.83E-01 2.33E-01 Basal
TABLE-US-00003 TABLE 2 Transcriptional, genomic and phenotypic
characteristics of cell lines in the panel. PIK3CA MYC
Transcriptional Doubling (3q26.32) (8q24.21) Cell Transcriptional
Subtype + ERBB Time GISTIC GISTIC Line Subtype 2 Status Culture
Media (hrs) Amplification Amplification 184A1 Non-malignant,
Non-malignant, MEGM .sup.a 63 ND ND Basal Basal 184B5
Non-malignant, Non-malignant, MEGM .sup.a 58 ND ND Basal Basal
600MPE Luminal Luminal DMEM + 10% FBS 101 No Amp No Amp AU565
Luminal ERBB2AMP RPMI + 10% FBS 38 Low Amp High Amp BT20 Basal
Basal DMEM + 10% FBS 62 Low Amp Low Amp BT474 Luminal ERBB2AMP RPMI
+ 10% FBS 91 Low Amp Low Amp BT483 Luminal Luminal RPMI + 10% FBS
141 Low Amp Low Amp BT549 Claudin-low Claudin-low RPMI + 10% FBS 25
No Amp Low Amp CAMA1 Luminal Luminal DMEM + 10% FBS 70 No Amp Low
Amp HCC1143 Basal Basal RPMI1640 + 10% FBS 59 No Amp Low Amp
HCC1187 Basal Basal RPMI1640 + 10% FBS 71 No Amp Low Amp HCC1395
Claudin-low Claudin-low RPMI1640 + 10% FBS 84 No Amp Low Amp
HCC1419 Luminal ERBB2AMP RPMI1640 + 10% FBS 170 No Amp High Amp
HCC1428 Luminal Luminal RPMI1640 + 10% FBS 88 Low Amp High Amp
HCC1599 Basal Basal RPMI1640 + 10% FBS ND Low Amp High Amp HCC1806
Basal Basal RPMI1640 + 10% FBS 37 Low Amp High Amp HCC1937 Basal
Basal RPMI1640 + 10% FBS 49 Low Amp High Amp HCC1954 Basal ERBB2AMP
RPMI1640 + 10% FBS 46 Low Amp High Amp HCC202 Luminal ERBB2AMP
RPMI1640 + 10% FBS 201 Low Amp Low Amp HCC2185 Luminal Luminal
RPMI1640 + 10% FBS 165 High Amp High Amp HCC2218 Luminal ERBB2AMP
RPMI1640 + 10% FBS ND No Amp Low Amp HCC3153 Basal Basal RPMI1640 +
10% FBS 59 Low Amp High Amp HCC38 Claudin-low Claudin-low RPMI1640
+ 10% FBS 53 Low Amp Low Amp HCC70 Basal Basal RPMI1640 + 10% FBS
73 Low Amp Low Amp HS578T Claudin-low Claudin-low DMEM + 10% FBS 38
Low Amp Low Amp LY2 Luminal Luminal DMEM + 10% FBS 53 No Amp High
Amp MCF10A Non-malignant, Non-malignant, DMEM/F12 + 5% HS + 27 ND
ND Basal Basal IHE + CholeraToxin .sup.b MCF10F Non-malignant,
Non-malignant, DMEM/F12 + 5% HS + 51 ND ND Basal Basal IHE +
CholeraToxin .sup.b MCF12A Non-malignant, Non-malignant, DMEM/F12 +
5% HS + 33 ND ND Basal Basal IHE + CholeraToxin .sup.b MCF7 Luminal
Luminal DMEM + 10% FBS 51 No Amp High Amp MDAMB134VI Luminal
Luminal DMEM + 20% FBS 107 ND ND MDAMB157 Claudin-low Claudin-low
DMEM + 10% FBS 67 No Amp Low Amp MDAMB175VII Luminal Luminal DMEM +
10% FBS 107 ND ND MDAMB231 Claudin-low Claudin-low DMEM + 10% FBS
25 No Amp No Amp MDAMB361 Luminal ERBB2AMP DMEM + 10% FBS 74 No Amp
Low Amp MDAMB415 Luminal Luminal DMEM + 10% FBS 85 Low Amp Low Amp
MDAMB436 Claudin-low Claudin-low DMEM + 10% FBS 63 Low Amp Low Amp
MDAMB453 Luminal Luminal DMEM + 10% FBS 60 Low Amp Low Amp MDAMB468
Basal Basal DMEM + 10% FBS 52 No Amp Low Amp SKBR3 Luminal ERBB2AMP
McCoy's + 10% FBS 56 Low Amp High Amp SUM102PT Basal Basal Serum
Free Ham's 115 No Amp Low Amp F12 + IHE .sup.f SUM1315MO2
Claudin-low Claudin-low Ham's F12 + 113 No Amp Low Amp 5% FBS + IE
.sup.d SUM149PT Basal Basal Ham's F12 + 34 ND ND 5% FBS + IH .sup.c
SUM159PT Claudin-low Claudin-low Ham's F12 + 22 No Amp High Amp 5%
FBS + IH .sup.c SUM185PE Luminal Luminal Ham's F12 + 93 No Amp Low
Amp 5% FBS + IH .sup.c SUM225CWN Luminal ERBB2AMP Ham's F12 + 73
Low Amp Low Amp 5% FBS + IH .sup.c SUM44PE Luminal Luminal Serum
Free Ham's 85 ND ND F12 + IH .sup.e SUM52PE Luminal Luminal Ham's
F12 + 53 Low Amp Low Amp 5% FBS + IH .sup.c T47D Luminal Luminal
RPMI1640 + 10% FBS 56 Low Amp Low Amp UACC812 Luminal ERBB2AMP DMEM
+ 10% FBS 99 No Amp Low Amp UACC893 Luminal Luminal DMEM + 10% FBS
153 ND ND ZR751 Luminal Luminal RPMI1640 + 10% FBS 68 No Amp Low
Amp ZR7530 Luminal Luminal RPMI1640 + 10% FBS 336 ND ND ZR75B
Luminal Luminal RPMI1640 + 10% FBS 63 No Amp Low Amp CCND1 ERBB2
AURKA CDKN2A PTEN Isogenic (11q13.2) (17q12) (20q13.2) (9p21.3)
(10q23.31) cell Cell GISTIC GISTIC GISTIC GISTIC GISTIC line Line
Amplification Amplification Amplification Deletion Deletion pair
184A1 ND ND ND ND ND na 184B5 ND ND ND ND ND na 600MPE High Amp Low
Amp No Amp Low Del No Del na AU565 No Amp High Amp High Amp Low Del
Low Del SKBR3 BT20 No Amp No Amp High Amp High Del Low Del na BT474
Low Amp High Amp High Amp Low Del No Del na BT483 Low Amp Low Amp
Low Amp Low Del Low Del na BT549 Low Amp No Amp Low Amp No Del No
Del na CAMA1 High Amp No Amp Low Amp No Del Low Del na HCC1143 High
Amp Low Amp Low Amp Low Del No Del na HCC1187 Low Amp No Amp No Amp
No Del No Del na HCC1395 Low Amp No Amp Low Amp High Del High Del
na HCC1419 Low Amp High Amp High Amp Low Del Low Del na HCC1428 Low
Amp No Amp High Amp No Del No Del na HCC1599 Low Amp Low Amp Low
Amp No Del No Del na HCC1806 Low Amp No Amp Low Amp High Del No Del
na HCC1937 Low Amp No Amp Low Amp Low Del High Del na HCC1954 High
Amp High Amp Low Amp Low Del Low Del na HCC202 No Amp High Amp Low
Amp No Del No Del na HCC2185 Low Amp No Amp Low Amp Low Del Low Del
na HCC2218 No Amp High Amp Low Amp No Del Low Del na HCC3153 Low
Amp Low Amp Low Amp No Del High Del na HCC38 No Amp Low Amp Low Amp
High Del Low Del na HCC70 No Amp No Amp Low Amp No Del Low Del na
HS578T No Amp No Amp Low Amp No Del No Del na LY2 Low Amp No Amp
High Amp High Del No Del MCF7 MCF10A ND ND ND ND ND na MCF10F ND ND
ND ND ND na MCF12A ND ND ND ND ND na MCF7 Low Amp No Amp High Amp
High Del No Del na MDAMB134VI ND ND ND ND ND na MDAMB157 No Amp No
Amp Low Amp No Del No Del na MDAMB175VII ND ND ND ND ND na MDAMB231
No Amp No Amp No Amp High Del No Del na MDAMB361 High Amp High Amp
High Amp Low Del No Del na MDAMB415 High Amp No Amp Low Amp No Del
Low Del na MDAMB436 No Amp No Amp Low Amp Low Del Low Del na
MDAMB453 High Amp Low Amp Low Amp Low Del No Del na MDAMB468 Low
Amp No Amp Low Amp No Del No Del na SKBR3 No Amp High Amp High Amp
Low Del Low Del na SUM102PT No Amp No Amp No Amp High Del No Del na
SUM1315MO2 No Amp No Amp No Amp High Del No Del na SUM149PT ND ND
ND ND ND na SUM159PT No Amp No Amp No Amp Low Del No Del na
SUM185PE No Amp No Amp Low Amp Low Del No Del na SUM225CWN Low Amp
High Amp Low Amp Low Del Low Del na SUM44PE ND ND ND ND ND na
SUM52PE Low Amp No Amp No Amp Low Del Low Del na T47D Low Amp Low
Amp Low Amp Low Del No Del na UACC812 Low Amp High Amp Low Amp Low
Del No Del na UACC893 ND ND ND ND ND na ZR751 High Amp No Amp Low
Amp No Del No Del na ZR7530 ND ND ND ND ND na ZR75B High Amp No Amp
Low Amp No Del No Del ZR751 .sup.a Clonetics MEBM (no Bi Carbonate)
+ Insulin(5 ug/ml) + Transferrin(5 ug/ml) + Hydrocortisone(0.5
ug/ml) + EGF(5 ng/ml) + Isoprorternol 10e-5M + Bovine Pituitary
Extracts 70 ug/ml) + Sodium Bicarbonate (1.176 bmg/ml) .sup.b
DMEM/F12 + 5% Horse serum + Insulin (10 ug/ml) + Hydrocortisone
(500 ng/ml) + EGF (20 ng/ml) + Cholera Toxin (100 ng/ml) .sup.c
Ham's F12 + 5% FBS + Insulin (5 ug/ml) + Hydrocortisone (1 ug/ml) +
HEPES (10 mM) .sup.d Ham's F12 + 5% FBS + Insulin (5 ug/ml) + HEPES
(10 mM) + EGF (10 ng/ml) .sup.e Ham's F12 + Insulin (5 ug/ml) +
HEPES (10 mM) + Hydrocortisone (1 ug/ml) + Ethanolamine(5 mM) +
Transferrin (5 ug/ml) + T3 (10 nM) + Sodium Selenite (50 nM) + BSA
(0.5 g/L) .sup.f Ham's F12 + Insulin (5 ug/ml) + HEPES (10 mM) +
Hydrocortisone (1 ug/ml) + Ethanolamine(5 mM) + Transferrin (5
ug/ml) + T3 (10 nM) + Sodium Selenite (50 nM) + BSA (0.5 g/L) +
EGF(10 ng/ml) .sup.g DMEM/F12 + Insulin (250 ng/ml) +
Hydrocortisone (1.4 nM) + Transferrin (10 ng/ml) + Sodium Selenite
(2.6 ng/ml) + Estradiol (100 nM) + Prolactin(5 ug/ml) + EGF(10
ng/ml) ND Not done na not applicable While we had no data to assign
ERBB2 status, literature suggests UACC893 and ZR7530 are ERBB2
amplified (PMID: 1674877, 688225)
TABLE-US-00004 TABLE 5 Subtype associations for all therapeutic
compounds. Basal/ Basal + Basal/ Basal + Claudin- Claudin-
ERBB2AMP/ Claudin- Claudin- ERBB2AMP/ low/ low/ not low/ low/ not
Luminal Luminal ERBB2AMP Luminal Luminal ERBB2AMP Compound Target
(q-val) (q-val) (q-val) (p-val) (p-val) (p-val) 17-AAG HSP90AA1
1.83E-01 1.10E-01 8.67E-02 4.54E-02 2.02E-02 1.36E-02 5-FdUR TYMS,
DNA, RNA 7.94E-01 5.74E-01 9.76E-01 5.94E-01 3.44E-01 9.47E-01 5-FU
TYMS, DNA, RNA 3.81E-01 3.53E-01 3.37E-01 1.60E-01 1.35E-01
1.28E-01 AG1024 IGF1R 4.51E-01 3.22E-01 8.46E-01 2.31E-01 1.09E-01
7.09E-01 AG1478 EGFR 9.34E-01 9.34E-01 2.60E-02 8.92E-01 8.78E-01
1.29E-03 AS-252424 PIK3CG 9.34E-01 8.79E-01 3.87E-01 8.73E-01
7.71E-01 1.67E-01 AZD6244 MAP2K1, MAP2K2 9.34E-01 8.03E-01 8.46E-01
8.57E-01 6.12E-01 7.06E-01 BEZ235 PIK3CA 4.33E-01 6.10E-01 6.89E-01
2.05E-01 3.68E-01 4.61E-01 BIBW 2992 EGFR, ERBB2 6.93E-01 8.08E-01
6.39E-03 4.74E-01 6.19E-01 2.02E-04 Bortezomib PSMD2, PSMB1,
9.34E-01 8.08E-01 8.79E-01 8.88E-01 6.28E-01 7.69E-01 PSMB5, PSMB2,
PSMD1 Bosutinib SRC 3.35E-01 1.82E-01 3.33E-01 1.25E-01 4.42E-02
1.22E-01 Carboplatin DNA cross-linker 3.22E-01 1.48E-01 5.06E-01
1.04E-01 3.27E-02 2.82E-01 CGC-11047 Polyamine analogue 6.51E-02
1.25E-01 8.08E-01 8.21E-03 2.47E-02 6.48E-01 CGC-11144 Polyamine
analogue 7.19E-01 6.80E-01 1.81E-01 5.09E-01 4.38E-01 4.32E-02
Cisplatin DNA cross-linker 8.45E-02 4.31E-02 8.52E-01 1.26E-02
3.37E-03 7.29E-01 CRT-11 TOP1 4.33E-01 9.83E-01 9.34E-01 2.09E-01
9.76E-01 8.88E-01 Docetaxel TUBB1, BCL2 8.67E-02 4.83E-02 8.44E-01
1.47E-02 4.32E-03 6.99E-01 Doxorubicin TOP2A 8.52E-01 7.75E-01
5.05E-01 7.28E-01 5.72E-01 2.73E-01 Epirubicin TOP2A 8.03E-01
8.12E-01 8.08E-01 6.11E-01 6.55E-01 6.44E-01 Erlotinib EGFR
9.48E-02 2.83E-01 2.33E-01 1.67E-02 8.28E-02 6.19E-02 Etoposide
TOP2A 3.34E-02 5.13E-02 8.89E-01 2.38E-03 5.77E-03 7.96E-01
Fascaplysin CDK4 4.83E-02 4.31E-02 3.70E-01 4.37E-03 3.50E-03
1.52E-01 Gefitinib EGFR 4.89E-01 3.35E-01 4.14E-03 2.62E-01
1.24E-01 1.12E-04 Geldanamycin HSP90AA1 9.86E-01 9.83E-01 2.02E-01
9.86E-01 9.78E-01 5.27E-02 Gemcitabine Pyrimidine 4.00E-01 6.20E-01
5.05E-01 1.80E-01 3.85E-01 2.75E-01 animetabolite Glycyl-H-1152
ROCK2 5.53E-01 3.53E-01 2.78E-01 3.26E-01 1.37E-01 7.71E-02
GSK1059615 PIK3CA 2.02E-01 1.03E-01 1.98E-01 5.18E-02 1.86E-02
5.01E-02 GSK1070916 AURKB, AURKC 5.13E-02 4.83E-02 4.82E-01
5.59E-03 4.57E-03 2.56E-01 GSK1120212 MAP2K1, MAP2K2 6.11E-01
3.91E-01 9.83E-01 3.71E-01 1.73E-01 9.72E-01 GSK1487371 PIK3CG
9.34E-01 9.02E-01 2.78E-01 8.82E-01 8.17E-01 7.56E-02 GSK1838705
IGF1R 6.89E-01 4.38E-01 5.23E-01 4.65E-01 2.13E-01 3.02E-01
GSK2119563 PIK3CA 2.85E-02 8.11E-03 8.67E-02 1.54E-03 3.01E-04
1.43E-02 GSK2126458 PIK3CA, PIK3CB, 1.27E-03 1.27E-03 8.67E-02
2.87E-05 2.41E-05 1.43E-02 PIK3CD, PIK3CG GSK461364 PLK1 3.21E-01
1.54E-01 7.87E-01 1.01E-01 3.53E-02 5.85E-01 GSK923295 CENPE
4.77E-01 3.16E-01 4.51E-01 2.52E-01 9.82E-02 2.28E-01 Ibandronate
FDPS 6.16E-01 5.06E-01 3.25E-01 3.80E-01 2.85E-01 1.14E-01 sodium
salt ICRF-193 TOP2BA, TOP2AB 4.33E-01 9.77E-01 8.08E-01 2.09E-01
9.55E-01 6.33E-01 Ispinesib Kinesin 6.93E-01 6.16E-01 9.76E-01
4.71E-01 3.80E-01 9.50E-01 Ixabepilone TUBB3 2.81E-01 1.22E-01
3.11E-01 7.96E-02 2.36E-02 9.54E-02 L-779450 BRAF 8.89E-01 9.34E-01
8.08E-01 7.97E-01 8.87E-01 6.35E-01 Lapatinib EGFR, ERBB2 7.23E-02
3.34E-02 2.26E-06 9.59E-03 2.29E-03 1.02E-08 LBH589 HDAC 5.14E-02
3.34E-02 3.22E-01 6.02E-03 2.41E-03 1.09E-01 Lestaurtinib FLT3,
NTRK1 4.00E-01 4.13E-01 5.41E-01 1.79E-01 1.93E-01 3.14E-01
Methotrexate DHFR 5.08E-01 9.76E-01 3.56E-01 2.88E-01 9.46E-01
1.40E-01 MLN4924 NAE1 7.01E-01 4.54E-01 9.81E-01 4.89E-01 2.35E-01
9.64E-01 NSC 663284 CDC25A, CDC25B, 5.53E-01 6.48E-01 7.64E-01
3.25E-01 4.09E-01 5.50E-01 CDC25C NU6102 CDK1, CCNB1 4.40E-01
5.06E-01 2.83E-01 2.18E-01 2.85E-01 8.28E-02 Nutlin 3a MDM2
5.19E-01 3.56E-01 7.68E-01 2.97E-01 1.42E-01 5.57E-01 Oxaliplatin
DNA cross-linker 7.00E-01 8.12E-01 3.89E-01 4.86E-01 6.59E-01
1.70E-01 Oxamflatin HDAC 7.22E-01 4.69E-01 8.79E-01 5.14E-01
2.45E-01 7.72E-01 Paclitaxel TUBB1, BCL2 6.21E-01 3.87E-01 7.95E-01
3.89E-01 1.67E-01 5.98E-01 PD 98059 MAP2K1, MAP2K2 8.08E-01
8.89E-01 9.25E-01 6.28E-01 7.95E-01 8.41E-01 PD173074 FGFR3
5.13E-02 3.68E-01 5.06E-01 5.54E-03 1.49E-01 2.78E-01 Pemetrexed
TYMS, DHFR, GART 4.44E-01 8.08E-01 3.70E-01 2.22E-01 6.42E-01
1.53E-01 Purvalanol A CDK1 3.22E-01 6.89E-01 9.38E-01 1.08E-01
4.50E-01 9.00E-01 Rapamycin MTOR 1.45E-02 8.11E-03 3.84E-01
6.52E-04 3.29E-04 1.63E-01 SB-3CT MMP2, MMP9 6.89E-01 6.89E-01
9.34E-01 4.62E-01 4.65E-01 8.79E-01 Sigma AKT1-2 AKT1, AKT2
1.17E-03 2.63E-04 1.29E-01 1.59E-05 2.37E-06 2.79E-02 inhibitor
Sorafenib KDR 8.79E-01 7.24E-01 6.58E-01 7.56E-01 5.18E-01 4.18E-01
Sunitinib Malate KDR 6.89E-01 4.40E-01 3.22E-01 4.54E-01 2.17E-01
1.10E-01 Tamoxifen ESR1 2.95E-01 1.29E-01 8.52E-01 8.91E-02
2.72E-02 7.24E-01 TCS 2312 CHEK1 3.28E-01 4.51E-01 5.73E-01
1.17E-01 2.31E-01 3.41E-01 dihydrochloride TCS JNK 5a MAPK9, MAPK10
8.24E-01 7.75E-01 8.89E-01 6.72E-01 5.71E-01 7.91E-01 Temsirolimus
MTOR 1.64E-01 7.25E-02 1.29E-01 3.84E-02 1.01E-02 2.77E-02 TGX-221
PIK3CB 4.10E-01 2.78E-01 4.09E-01 1.90E-01 7.76E-02 1.88E-01
Topotecan TOP1 8.30E-01 8.08E-01 7.71E-01 6.80E-01 6.45E-01
5.62E-01 TPCA-1 IKBKB 3.22E-01 1.54E-01 1.29E-01 1.12E-01 3.50E-02
2.64E-02 Trichostatin A HDAC 1.22E-01 5.13E-02 7.10E-01 2.30E-02
5.78E-03 4.99E-01 Triciribine AKT, ZNF217 8.67E-02 5.91E-02
3.56E-01 1.48E-02 7.18E-03 1.43E-01 Vinorelbine TUBB 3.22E-01
3.32E-01 9.34E-01 1.05E-01 1.20E-01 8.72E-01 Vorinostat HDAC
7.23E-02 3.34E-02 6.89E-01 9.77E-03 2.34E-03 4.62E-01 VX-680 AURKA,
AURKB, 2.95E-01 4.02E-01 7.77E-02 8.80E-02 1.83E-01 1.12E-02 AURKC
XRP44X ELK3 8.79E-01 6.98E-01 9.02E-01 7.71E-01 4.81E-01 8.13E-01
ZM 447439 AURKA 8.32E-01 6.80E-01 8.52E-01 6.86E-01 4.38E-01
7.29E-01
* * * * *
References